A single-objective EPR based model for creep index of soft clays considering L2 regularization

Yin Fu Jin, Zhen Yu Yin, Wan Huan Zhou, Jian Hua Yin, Jian Fu Shao

Research output: Journal article publicationJournal articleAcademic researchpeer-review

55 Citations (Scopus)


A correlation of creep index (Cα) with high performance is highly recommended for soft clay engineering practice. Current empirical correlations are only acceptable for a few clays. This paper aims to propose a robust and effective evolutionary polynomial regression (EPR) model for Cα of clay. First, a database covering various clays is formed, in which 120 data are randomly selected for training and the remaining data are used for testing. To avoid overfitting, a novel EPR procedure using a newly enhanced differential evolution (DE) algorithm is proposed with two enhancements: (1) a new fitness function is proposed using the structural risk minimization (SRM) with L2 regularization that penalizes polynomial complexity, and (2) an adaptive process for selecting the combination of involved variables and size of polynomial terms is incorporated. By comparing the predictive ability, model complexity, robustness and monotonicity, the EPR formulation for Cα involving clay content, plasticity index and void ratio with three terms is selected as the optimal model. A parametric study is then conducted to assess the importance of each input in the proposed model. All results demonstrate that the proposed model of Cα is simple, robust, and reliable for applications in engineering practice.

Original languageEnglish
Pages (from-to)242-255
Number of pages14
JournalEngineering Geology
Publication statusPublished - 8 Jan 2019


  • Atterberg limits
  • Creep
  • Evolutionary polynomial regression
  • Model selection
  • Optimization
  • Soft clays

ASJC Scopus subject areas

  • Geotechnical Engineering and Engineering Geology
  • Geology


Dive into the research topics of 'A single-objective EPR based model for creep index of soft clays considering L2 regularization'. Together they form a unique fingerprint.

Cite this